Comparison of tumor‑targeting properties of directly and indirectly radioiodinated designed ankyrin repeat protein (DARPin) G3 variants for molecular imaging of HER2

Abstract

Evaluation of human epidermal growth factor receptor 2 (HER2) expression levels in breast and gastroesophageal cancer is used for the stratification of patients for HER2‑targeting therapies. The use of radionuclide molecular imaging may facilitate such evaluation in a non‑invasive way. Designed ankyrin repeat proteins (DARPins) are engineered scaffold proteins with high potential as probes for radionuclide molecular imaging. DARPin G3 binds with high affinity to HER2 and may be used to visualize this important therapeutic target. Studies on other engineered scaffold proteins have demonstrated that selection of the optimal labeling approach improves the sensitivity and specificity of radionuclide imaging. The present study compared two methods of labeling G3, direct and indirect radioiodination, to select an approach providing the best imaging contrast. G3‑H6 was labeled with iodine‑124, iodine‑125 and iodine‑131 using a direct method. A novel construct bearing a C‑terminal cysteine, G3‑GGGC, was site‑specifically labeled using [125I]I‑iodo‑[(4‑hydroxyphenyl)ethyl]maleimide (HPEM). The two radiolabeled G3 variants preserved binding specificity and high affinity to HER2‑expressing cells. The specificity of tumor targeting in vivo was demonstrated. Biodistribution comparison of [131I]I‑G3‑H6 and [125I]I‑HPEM‑G3‑GGGC in mice, bearing HER2‑expressing SKOV3 xenografts, demonstrated an appreciable contribution of hepatobiliary excretion to the clearance of [125I]I‑HPEM‑G3‑GGGC and a decreased tumor uptake compared to [131I]I‑G3‑H6. The direct label provided higher tumor‑to‑blood and tumor‑to‑organ ratios compared with the indirect label at 4 h post‑injection. The feasibility of high contrast PET/CT imaging of HER2 expression in SKOV3 xenografts in mice using [124I]I‑G3‑H6 was demonstrated. In conclusion, direct radioiodination is the preferable approach for labeling DARPin G3 with iodine‑123 and iodine‑124 for clinical single photon emission computed tomography and positron emission tomography imaging.

Keywords: DARPin; HER2; imaging; radionuclide; iodine; radioiodination.

Figure 1

Site-specific radioiodination of G3-GGGC is a one-pot reaction achieved in two steps without intermediate purification. In the first step, HPEM is iodinated using chloramine T; in the second step, HPEM is attached to the protein using maleimide-cysteine coupling. HPEM, (4-hydroxyphenyl)ethyl) maleimide; RT, room temperature.

Figure 2

Specificity of [¹²⁵I]I-G3-H₆ and [¹²⁵I]I-HPEM-G3-GGGC binding to cancer cell lines with high (SKOV3, BT474) and low (A431, DU145) levels of HER2 expression in vitro. Radiolabeled compounds were added at a 1 nM concentration; 100-fold molar excess of non-labeled G3-H₆ was added to the blocked groups. Data are presented as the mean ± standard deviation from three samples. The uptake of each compound in the non-blocked group was compared with the uptake in the blocked group in each cell line individually. ^*P<0.05. HPEM, (4-hydroxyphenyl)ethyl) maleimide; HER2, human epidermal growth factor receptor 2.

Figure 3

In vitro characterization of radio-iodinated G3 variants. Cellular processing of (A) [¹²⁵I]I-HPEM-G3-GGGC and (B) [¹²⁵I]I-G3-H₆ by human epidermal growth factor receptor 2-expressing SKOV3 cells during continuous incubation over 24 h. Data are presented as the mean ± standard deviation from three samples; error bars may not be visible when they are smaller than symbols. HPEM, (4-hydroxyphenyl)ethyl) maleimide.

Figure 4

Comparative biodistribution of [¹²⁵I]I-HPEM-G3-GGGC and [¹³¹I]I-G3-H₆ at 4 h pi in BALB/C nu/nu mice bearing human epidermal growth factor receptor 2-expressing SKOV3 xenografts. Data are presented as the mean ± standard deviation from four mice. An asterisk marks a signifi-cant difference between values. ^*P<0.05. ^aData for intestines with content and carcass are presented as %ID per whole sample. HPEM, (4-hydroxyphenyl) ethyl) maleimide; sal. gland, salivary gland; ID, injected dose.

Figure 5

Comparison of tumor-to-organ ratios for [¹²⁵I]I-HPEM-G3-GGGC and [¹³¹I]I-G3-H₆ at 4 h pi in BALB/C nu/nu mice bearing SKOV3 xenografts. Data are presented as the mean ± standard deviation from four mice. ^*P<0.05. HPEM, (4-hydroxyphenyl)ethyl) maleimide; sal. gland, salivary gland.

Figure 6

Tumor-targeting specificity of (A) [¹²⁵I]I-HPEM-G3-GGGC and (B) [¹³¹I]I-G3-H₆ in SKOV3 xenografts (n=4; high HER2 expression) and A431 xenografts (n=3; low HER2 expression). Data are presented as the mean ± standard deviation. ^*P<0.05. HPEM, (4-hydroxyphenyl)ethyl) maleimide; ID, injected dose; HER2, human epidermal growth factor receptor 2.

Figure 7

Imaging of HER2 expression in mice bearing SKOV3 xenografts at 1, 2 and 4 h. Micro-single photon emission computed tomography/computed tomography imaging of human epidermal growth factor receptor 2 expression in BALB/C nu/nu mice bearing SKOV3 xenograft at (A) 1, (B) 2 and (C) 4 h post-injection using [¹²⁵I]I-G3-H₆. The uptake of radiocatabolites by Na/I-symporters was blocked by addition of KI in drinking water. The area corresponding to the activity in urinary bladder was removed from the 1 and 2 h images during reconstruction to enable better visualization of tumor uptake. S, stomach; K, kidneys; T, tumor.

Figure 8

Comparative imaging of high (SKOV3) and low (A431) HER2-expressing xenografts in mice. Micro-single photon emission computed tomography/computed tomography imaging of human epidermal growth factor receptor 2 expression in BALB/C nu/nu mice bearing (A) SKOV3 and (B) A431 xenografts at 4 h post-injection using [¹²⁵I]I-G3-H₆. The uptake of radiocatabolites by Na/I-symporters was blocked by addition of KI in drinking water. S, stomach; K, kidneys; T, tumor.

Figure 9

MicroPET/computed tomography imaging of human epidermal growth factor receptor 2 expression in BALB/C nu/nu mouse bearing SKOV3 xenograft at 4 h post-injection using [¹²⁴I]I-G3-H₆. The uptake of radiocatab-olites by Na/I-symporters was blocked by addition of KI in drinking water. The area corresponding to the activity in urinary bladder was removed from the PET image during reconstruction to enable better visualization of tumor uptake. S, stomach; T, tumor; PET, positron emission tomography.